INDEX

Explanations

references to political affiliations or ideologies, particularly related to the left side of the political spectrum

oai_token-act-pair · gpt-3.5-turbo

references to the political left

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.3.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Frey

-0.70

actionDate

-0.68

glomer

-0.67

alez

-0.67

MAT

-0.66

TAG

-0.65

 Mehran

-0.64

LIA

-0.64

oun

-0.64

andise

-0.64

POSITIVE LOGITS

wing

1.08

overs

1.03

ward

0.88

 wing

0.87

 hemisphere

0.83

 orthodoxy

0.81

 fringe

0.78

 populism

0.78

 Syri

0.77

orthodox

0.75

Activations Density 0.032%

references to political affiliations or ideologies, particularly related to the left side of the political spectrum

references to the political left

No Comments

No Known Activations